To take raw data into a report colleagues can use, prepare and model the data in Power BI Desktop, build report pages, publish the report and semantic model to a Power BI service workspace, then grant access through the right route—often an app for readers. The key is to treat data preparation, modeling, report design, publishing, and distribution as connected but distinct stages.
This guide follows Microsoft’s end-to-end Power BI tutorial. The exact connection, refresh, gateway, and access requirements depend on your data source and organization’s setup.
1. Connect to the data you need
Start in Power BI Desktop and connect to the source that contains the data for the report. Before building, identify who owns the source, how it is authenticated, and whether Power BI Service will be able to reach it for later updates. Power BI supports multiple cloud, on-premises, and other connection scenarios; the right setup depends on where the data lives and how your organization manages access. See Microsoft’s Power BI documentation for the range of supported scenarios.
- Confirm the source and the person or team responsible for it.
- Check that you have the credentials and permissions needed to connect.
- Plan how the service will reach the source after publication; a connection that works only on your computer may not be sufficient.
2. Prepare and shape the data
Use Power Query to select the needed data, correct data types, clean or reshape columns, and combine tables where appropriate. The goal is to load data with a clear structure and meaning—not to rely on chart formatting to compensate for incorrect or ambiguous source data.
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Scan for outdated or missing drivers - takes under a minuteDriver Scan →Repair Windows errors before they cause bigger problemsFix Now →#1 Best Overall
Keep the grain of each table in view: establish what one row represents before combining or summarizing data. Microsoft separates data transformation, shaping, and modeling into distinct learning areas in its transformation and modeling documentation.
3. Build a reusable semantic model
After shaping the data, establish the relationships and calculations that reports will rely on. A semantic model is the reusable data and calculation layer; the report is the analysis and visual presentation built on top of it. Defining relationships and reusable measures deliberately helps keep calculations consistent across report pages and, where appropriate, other reports.
Rank #2
Model design is a separate phase from choosing charts. Microsoft’s workflow tutorial includes building a semantic model intended for reuse.
4. Create report pages for the audience’s questions
Use Power BI Desktop to build report pages and visuals that answer specific questions for the intended audience. Keep the model’s calculations and relationships distinct from the report’s layout: the same semantic model can support analysis presented in different report pages or reports.
Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Fix the driver behind crashes, sound loss and screen glitches3Clear out junk files and repair common Windows errors5. Publish to a workspace
Publishing from Power BI Desktop sends the report and semantic model to a Power BI Service workspace. When publishing a PBIX file from a local drive, the local file remains on that drive; the service receives its own report and model artifacts. Microsoft explains this behavior in Get data from Power BI Desktop files.
For a local file source, the online model may need to be refreshed to stay current with the source. Publishing is therefore a handoff to the service, not a guarantee that the service can always reach or update every source automatically.
Rank #4
6. Choose the right refresh and connection setup
“Refresh” can mean different operations in Power BI, including refreshing model data, query caches, tiles, or report visuals. The relevant behavior depends on the storage mode and where the data is hosted. Microsoft distinguishes these operations in its data refresh documentation.
Import and DirectQuery
| Storage mode | What happens when data is updated | Practical implication |
|---|---|---|
| Import | A data refresh loads data from the source into the semantic model. | Plan model refreshes to keep the service copy current. A refreshed model does not necessarily redraw a report that is already open; report visuals may also need refreshing. |
| DirectQuery | Report visuals query the source when they are refreshed. | Scheduled semantic-model data refresh does not apply in the same way as it does for imported data. Consider source reachability and responsiveness when setting expectations. |
Cloud and on-premises sources
Cloud sources use cloud connections. Microsoft documents personal and shareable connection patterns, and notes that the shared-connection tasks on its cloud data sources page require a Power BI Pro or Premium Per User license, or a workspace in Premium/Fabric capacity. Licensing and role rules can change, so confirm current requirements for your organization and workspace.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
If the service cannot directly reach a source on a private or on-premises network, configure a gateway for service refresh. The server and database names in the Desktop connection must match those in the gateway definition, and a model with multiple sources needs each source configured. Grant access to gateway data-source connections only to trusted accounts that need it. See Microsoft’s instructions for adding or removing a gateway data source.
When the source schema changes
A renamed or removed source table or column can cause service refresh to fail. Microsoft notes that schema refresh is not performed in the service in the same way; refresh the schema in Desktop and republish when needed. Schema changes can also affect visuals, DAX expressions, security rules, and relationships, so check those dependencies as part of the update.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.7. Distribute the report with the right access
The workspace is where the report and semantic model live. An app packages content for its audience, making it a useful route for sharing with report consumers. Update the app after changing its content if you want those changes to reach its audience. Microsoft’s end-to-end tutorial covers publishing and distributing through an app.
Choose permissions according to what each person needs. Viewing an app or report, collaborating in a workspace, and reusing the underlying semantic model are different activities. Build permission on the semantic model allows consumers to reuse it in their own reports or Excel analyses; it is separate from app audience access. Publishing and app consumption can also depend on workspace roles, licenses, audience settings, and capacity. Check Microsoft’s current access and licensing guidance before deployment.
Recommended Free Tools
Quick Recap
Before you share: a practical checklist
- Source data is prepared with appropriate types, structure, and table grain.
- Relationships and reusable measures are in the semantic model, rather than being improvised separately in each visual.
- The service can reach each source, with cloud connections or a correctly configured gateway as appropriate.
- The refresh approach matches the storage mode and the audience’s freshness needs.
- Report consumers have the access they need; grant Build permission only when model reuse is intended.
- The app is updated when you want published changes to be visible to its audience.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
